CCTA-derived strain analysis in detection of regional myocardial dysfunction in coronary artery disease patients with preserved left ventricular ejection fraction: A feasibility study

Abstract

OBJECTIVES:

To evaluate the feasibility of using coronary computed tomography angiography (CCTA)-derived strain to detect regional myocardial dysfunction in coronary artery disease (CAD) patients with normal left ventricular ejection fraction (LVEF).

METHODS:

A total of 1,580 segments from 101 patients who underwent stressed CT myocardial perfusion imaging (CT-MPI) and CCTA were retrospectively enrolled in this study. The CT-derived global and segmental strain values were evaluated using the feature tracking technique. Segments with myocardial blood flow (MBF) < 125 ml/min/100 ml and 95 ml/min/100 ml were categorized as ischemic and infarcted, respectively.

RESULTS:

Segmental radial strain (SRS) and segmental circumferential strain (SCS) in the abnormal segments (including all segments with MBF < 125 ml/min/100 ml) were significantly lower than those in the normal segments (14.81±8.65% vs 17.17±9.13%, p < 0.001; –10.21±5.79% vs –11.86±4.52%, p < 0.001, respectively). SRS and SCS values in infarcted segments were significantly impaired compared with the ischemic segments (12.43±8.03% vs. 15.32±8.71%, p = 0.038; –7.72±5.91% vs. –10.67±5.66%, p = 0.010, respectively). The AUCs for SRS and SCS in detecting infarcted segments were 0.622 and 0.698, respectively (p < 0.05).

CONCLUSIONS:

It is feasible for using CCTA-derived strain parameters to detect regional myocardial dysfunction in CAD patients with preserved LVEF. Segmental radial and circumferential strain have the potential ability to distinguish myocardial ischemia from infarction, and normal from ischemic myocardium.

Keywords

Strain coronary computed tomography angiography myocardial blood flow myocardial ischemia myocardial infarction

1 Introduction

Coronary artery disease (CAD), which is characterized by atherosclerotic plaque formation in the coronary arteries, is considered the cause of myocardial ischemia and myocardial infarction [1]. Fractional flow reserve (FFR), single photon emission computed tomography (SPECT) and cardiac magnetic resonance (CMR) are used in current clinical practice to detect myocardial ischemia with different limitations, such as high price, increased radiation and additional requirements for invasive or semi-invasive procedures etc. Coronary computed tomography angiography (CCTA) has been widely used as an accurate method for assessing anatomic stenosis [2]. However, anatomic stenosis and myocardial ischemia may mismatch in CAD patients [3]. Stress dynamic computed tomography myocardial perfusion imaging (CT-MPI) combined with CCTA has been recognized as an ideal one-stop shop imaging modality and can assess both myocardial blood flow (MBF) and stenostic degree of coronary artery [4]. However, stress CT-MPI is subject to several limitations, such as increase of radiation dose, and the use of vasodilator stress agent, increasing the risk of examination [5].

Myocardial ischemia leads to regional myocardial dysfunction in the specific territory supplied by a stenotic coronary artery [6]. Strain analysis is a quantitative tool used in the early evaluation of regional myocardial dysfunction. Strain measured by CMR feature tracking has been explored in the literature in ischemic heart disease. Previous studies showed that myocardial strain parameters based on CMR were reduced in ischemic areas, in comparison to non-ischemic areas [7, 8]. Retrospective electrocardiogram (ECG) gating of CCTA with multi-phase reconstruction allows for three-dimensional (3D) strain analysis by the feature tracking technique [9]. CCTA-derived global and segmental strain has shown good agreement with strain using feature tracking CMR in cardiomyopathy diseases [10, 11]. However, to the best of the knowledge of the authors, few studies have been conducted on the capability of CT-derived strain to distinguish myocardial ischemia from infarction, especially in CAD patients with preserved left ventricular ejection fraction (LVEF). Therefore, the purpose of our study was to primarily investigate the feasibility of using strain parameters derived from CCTA to detect the regional myocardial dysfunction in CAD patients.

2 Materials and methods

2.1 Patients

The study enrolled retrospectively 109 patients with clinically suspected CAD from June 2018 to November 2019 according to inclusion criteria of aged 18 years or older, preserved EF (LVEF ≥ 50% measured by echocardiography) and successfully undergoing stress dynamic CT-MPI combined with CCTA. The exclusion criteria were as follows: 1) inadequate quality of CT-MPI or CCTA images (n = 4); 2) missing MBF information(n = 2); 3) failing to assess strain parameters (n = 2). In the enrolled 101 patients, thirty-six left ventricular myocardial segments were excluded due to poor tracking. Finally, 1580 left ventricular myocardial segments were included in the final analysis. The radiation dose was calculated from the dose-length product in a dose report.

This study was approved by our institutional review board and the informed consent requirement was waived due to its retrospective nature.

2.2 CT protocol and post processing

CT examinations were performed using a 96-row detector third-generation dual-source CT (SOMATOM Force, Siemens Healthineers, Forchheim, Germany). Double intravenous accesses in the forearms were implanted. Intravenous adenosine-triphosphate (ATP) (160μg/kg/min) was then administered 3 min prior to stress CT-MPI scan. During the stress dynamic MPI scan, a bolus of contrast medium (50 ml; iopromide (Bayer Healthcare, Berlin, Germany)) was injected into the antecubital vein at a rate of 6 mL/s, and 40 mL of normal saline were flushed to wash out the contrast medium from the superior vena cava and right ventricle. The entire LV was covered under the shuttle mode (i.e., the table was moving back and forth between acquisitions) [12]. Fixed tube voltage of 70 kV and attenuation-based tube current modulation were enabled during CT-MPI. A prospective ECG gating with an acquisition window of 250 ms, was applied to ensure myocardial perfusion captured at the end-systolic phase to reduce the beam-hardening artifact. Depending on the heart rate, a scan was performed every second or third heart cycle, resulting in a series of 10–15 acquired phases in total. Slice thickness of the reconstructed data was 3 mm.

An interval of 5 min between the MPI and CCTA provides optimal contrast and ATP washout. Retrospective ECG-gating CCTA with tube current modulation during the cardiac cycle was performed, using a bolus tracking technique and the region of interest (ROI) located in the ascending aorta [13]. The CCTA scanning parameters were as follows: the pitch, which adapted to the heart rate, tube voltage, and current selected automatically by CARE Dose 4D mode with reference to tube current–time product of 400 mAs, was 0.2 –0.39. Twenty phases were reconstructed in 5% steps of the R-R interval within the full window. Finally, the images were reconstructed using an iterative algorithm (ADMIRE, Siemens, Healthcare, strength set at a level of 3) with section thickness of 0.75mm for coronary stenosis analysis.

2.3 Image analysis

Perfusion images were reconstructed using a dedicated kernel for reduction of iodine beam-hardening artifacts and analyzed using a CT-MPI software package (SyngoVia, version VB10A, VPCT, Siemens Healthineers, Germany). Quantification of the MBF was performed using a hybrid deconvolution model [14]. According to the 17-segment model [15] and excluding the apical segment, the ROI was manually placed on the short-axis (SAX) view, and the MBF was collected on a segment base. The ROI was drawn in order to cover the entire area of the suspected perfusion defects within the segment. According to previous studies [16 –18], segments were recognized to be myocardial ischemia (MBF < 125 ml/min/100 ml) and myocardial infarction (MBF < 95 ml/min/100 ml), respectively, in this study. Subsequently, the patients were divided into normal, ischemic and infarcted group, according to the segment with lowest MBF. The end-diastolic volume index (EDVi), end-systolic volume index (ESVi), stroke volume index (SVi), LVEF, left ventricular mass index (LVMi), and cardiac index (CI) were obtained by the cardiac function analysis software (SyngoVia, version VB10A, Cardiac Function, Siemens Healthineers, Germany).

A commercial software package (Circle CVI 42, Circle Cardiovascular Imaging Inc., Canada) was used to analyze the global and segmental strain parameters. Twenty phases of CCTA images were imported into CVI 42 and reconstructed for LV SAX slices, and long axis (LAX) slices (two-, three-, and four-chamber views) based on maximal intensity projection (MIP) images. Endocardial and epicardial borders and interventricular septum insertion points were semi-automatically defined at the end-diastolic phase. The software then automatically propagated the delineations over the entire heart cycle. Finally, global and segmental strain parameters were calculated by feature tracking in the radial, circumferential and longitudinal directions, based on the 3D results (Fig. 1).

Fig. 1

Example of CT-MPI and strain analysis. Upper row shows images of a 66-year-old male patient with normal myocardial perfusion in CT-MPI, (A) whose lowest MBF was 134 ml/min/100 ml. No significant reduction in SRS) (B), SCS(C) and SLS(D) was detected. Lower row shows images of a 77-year-old male ischemic patient with infarcted myocardium, (E) whose lowest MBF was 95ml/min/100 ml in the inferior-septal segment at basal level. There was reduction of SRS(F), SCS(G) and SLS(H) in the inferior-septal myocardium compared to the normal patients. SRS, segmental radial strain; SCS, segmental circumferential strain; SLS, segmental longitudinal strain.

All image analysis was performed by Z.M.M. (with 2 years of cardiac imaging experience). The extracted features were shown in the Table 1. Then twenty patients were randomly selected to be evaluated the inter (Z.M.M. and G.Y.J.) and intra (4-week interval by Z.M.M.) reproducibility analysis.

Table 1

Extracted features

Images	Features
CCTA
Cardiac function	EDVi (end-diastolic volume index)
SyngoVia, version VB10A, Cardiac Function, Siemens Healthineers, Germany	ESVi (end-systolic volume index)
	SVi (stroke volume index)
	LVEF (left ventricular ejection fraction)
	LVMi (left ventricular mass index)
	CI (cardiac index)
Strain values	GRS (global radial strain)
Circle CVI 42, Circle Cardiovascular Imaging Inc., Canada	GCS (global circumferential strain)
	GLS (global longitudinal strain)
	SRS (segmental radial strain)
	SCS (segmental circumferential strain)
	SLS (segmental longitudinal strain)
CT-MPI
SyngoVia, version VB10A, VPCT, Siemens Healthineers, Germany	MBF (myocardial blood flow)

CCTA, coronary computed tomography angiography; CT-MPI, computed tomography myocardial perfusion imaging.

2.4 Statistical analysis

All continuous data were checked for normality using the Kolmogorov-Smirnov test. Continuous variables are expressed as mean±standard deviation, while categorical variables as frequency with percentage. Global strain and function parameters were compared among normal, ischemic and infarcted patients by one-way analysis of variance (ANOVA). Segment strain parameters were compared using ANOVA for comparing difference of regional function. Two-by-two comparison of continuous variables was used for the LSD (L) method. Independent sample Student’s t-test was used to compare segmental strain values between normal and abnormal segments, including all MBF < 125 ml/min/100 ml segments. Receiver operating characteristic (ROC) curve analysis and logistic regression analysis were performed at segment-level to determine the association between the strain values and infarcted segments, with the former expressed as the area under the curve (AUC) and the latter as odds ratio (OR). Reproducibility between observers in strain analysis was evaluated by intra-class correlation coefficient (ICC). Statistical significance was set at p≤0.05 in all tests. Statistical analyses were carried out using SPSS software (SPSS 21.0 for Windows, IBM, USA).

3 Results

3.1 Study population and segments baseline characteristics

Stress CT-MPI detected 51 infarcted, 236 ischemic and 1293 normal segments (Fig. 2). The patient characteristics are displayed in Table 2. Fifteen patients had severe stenosis, with at least one lesion lumen area stenosis > 70%. Eighteen patients had moderate stenosis, with at least one lesion lumen area stenosis between 50 –70%. Twenty-five patients underwent coronary stent implantation and four underwent coronary artery bypass grafting. The radiation dose was 3.48±0.52 mSv and 4.56±1.99 mSv for MPI and CCTA scan, respectively.

Fig. 2

Study flow chart. CT-MPI, computed tomography myocardial perfusion imaging; CCTA, coronary computed tomography angiography; MBF, myocardial blood flow.

Table 2

Patient characteristics

Number of patients	101
Age (years)	60±12
Males (% of total)	73(72%)
Body mass index (kg/m²)	24.7±2.6
Risk factors (number [% ])
Diabetes mellitus	20(20%)
Hypertension	58(57%)
Hypercholesterolemia	39(39%)
Smoking	29(29%)
Drinking	28(27%)
Average Heart rate(bpm)
Stress MPI	90.6±13.6
CCTA	72.5±11.0

Data are presented as the mean±standard deviation or the number (%) of subjects.

3.2 A comparison of cardiac function and strain values at patient level

Out of the 101 patients, 50 were characterized as normal, 33 as ischemic and 18 as infarcted. In cardiac function analysis, significant difference of LVMi was found among normal, ischemic and infarcted patients (p = 0.032). LVMi in the infarcted patients was significantly higher than that in normal patients (65.61±11.12 g/m² vs. 58.46±9.77 g/m², p = 0.009). Meanwhile, no significant differences were detected in other cardiac function parameters and global strain values in three directions among these three groups (Table 3).

Table 3
A comparison of cardiac function and global strain values among the normal, ischemic, and infarcted patients

Normal patient group (n = 50) Ischemic patient group (n = 33) Infarcted patient group (n = 18) P values

LVEF (%) 74.94±6.15 77.36±5.86 70.11±12.74 0.056

EDVi (mL/m²) 79.74±19.19 76.37±13.05 79.78±21.07 0.683

ESVi (mL/m²) 21.89±13.21 16.43±6.44 19.02±11.29 0.112

SVi (mL/m²) 57.94±11.58 59.75±9.72 54.11±9.08 0.197

CI (L/min/m²) 4.92±1.11 4.73±1.34 4.18±0.75 0.069

LVMi (g/m²) 58.46±9.77 60.39±8.27 65.61±11.12a 0.032*

GRS (%) 15.99±5.27 15.56±5.76 13.17±4.61 0.157

GCS (%) – 10.82±2.91 – 10.57±3.01 – 9.79±3.41 0.467

GLS (%) – 6.58±2.32 – 7.24±2.59 – 7.29±2.60 0.376

	Normal patient group (n = 50)	Ischemic patient group (n = 33)	Infarcted patient group (n = 18)	P values
LVEF (%)	74.94±6.15	77.36±5.86	70.11±12.74	0.056
EDVi (mL/m²)	79.74±19.19	76.37±13.05	79.78±21.07	0.683
ESVi (mL/m²)	21.89±13.21	16.43±6.44	19.02±11.29	0.112
SVi (mL/m²)	57.94±11.58	59.75±9.72	54.11±9.08	0.197
CI (L/min/m²)	4.92±1.11	4.73±1.34	4.18±0.75	0.069
LVMi (g/m²)	58.46±9.77	60.39±8.27	65.61±11.12a	0.032*
GRS (%)	15.99±5.27	15.56±5.76	13.17±4.61	0.157
GCS (%)	– 10.82±2.91	– 10.57±3.01	– 9.79±3.41	0.467
GLS (%)	– 6.58±2.32	– 7.24±2.59	– 7.29±2.60	0.376

Data are presented as the mean±standard deviation. LVEF, left ventricular ejection fraction; EDVi, end-diastolic volume index; ESVi, end-systolic volume index; SVi, stroke volume index; LVMi, left ventricular mass index; GRS, global radial strain; GCS, global circumferential strain; GLS, global longitudinal strain. ^*p < 0.05, a p < 0.05 compared with the normal patients.

3.3 A comparison of strain values at segment level

Segmental radial strain (SRS) and segmental circumferential strain (SCS) in abnormal segments were significantly lower than those in normal segments (14.81±8.65% vs 17.17±9.13%, p < 0.001; –10.21±5.79% vs –11.86±4.52%, p < 0.001). On the other hand, no significant difference was found in the segmental longitudinal strain (SLS) values between normal and abnormal segments (Fig. 3). SRS values were significantly impaired in both ischemic and infarcted segments, compared with the normal segments (15.32±8.71% vs 17.17±9.13%, p = 0.004; 12.43±8.03% vs 17.17±9.13%, p < 0.001). The same results were also found in the SCS values among normal, ischemic and infarcted segments (–10.67±5.66% vs –11.86±4.52%, p = 0.008; –7.72±5.91% vs –11.86±4.52%, p < 0.001). Moreover, infarcted segments had further reduction in SRS and SCS compared with ischemic segments at a significant level (12.43±8.03% vs. 15.32±8.71%, p = 0.038; –7.72±5.91% vs. –10.67±5.66%, p = 0.010). Finally, no significant difference in the SLS values was detected among normal, ischemic and infarcted segments (Table 4).

Fig. 3

Comparison of SRS (A), SCS (B) and SLS (C) in the normal and abnormal segments. SRS was significantly impaired in the abnormal segments, compared with the normal segments (A). The abnormal segments had significantly decreased SCS, compared with the normal segments (B). No significant difference was found between these two groups, with regards to SLS (C). SRS, segmental radial strain; SCS, segmental circumferential strain; SLS, segmental longitudinal strain. ^*p < 0.05.

Table 4

A comparison of strain values among the normal, ischemic, and infarcted segments

	Normal segments (n = 1293)	Ischemic segments (n = 236)	Infarcted segments (n = 51)	p
SRS (%)	17.17±9.13	15.32±8.71a	12.43±8.03 a,b	< 0.001*
SCS (%)	– 11.86±4.52	– 10.67±5.66a	– 7.72±5.91 a,b	< 0.001*
SLS (%)	– 6.98±2.85	– 7.25±3.21	– 6.97±2.94	0.421

SRS, segmental radial strain; SCS, segmental circumferential strain; SLS, segmental longitudinal strain. ^*p < 0.05, a p < 0.05 compared with the normal segments, b p < 0.05 compared with the ischemic segments.

3.4 Diagnostic capability of segment strain for infarcted segments

The results of the ROC analysis and logistic regression using strain parameters to predict infarcted segments from non-infarcted segments were illustrated in Table 5. Univariate logistic regression analysis showed that SRS and SCS were significantly associated with infarcted segments, while in the multivariate logistic regression, SCS was independently associated with infarcted segments. The AUCs for SRS and SCS in detecting infarcted segments were 0.622 and 0.698, respectively (p < 0.05) (Fig. 4). A cut-off of –10.73% for SCS allowed the differentiation of infarction segments with sensitivity 72.7%, specificity 57.5% and accuracy 57.9%.

Table 5
Results of the ROC analysis and logistic regression analysis demonstrating the association between strain measurements and infarcted segments

ROC analysis Univariate logistic regression Multivariate logistic regression

Cut-off AUC (95% CI) p OR (95% CI) p OR (95% CI) p

SRS 15.09 0.622(0.545, 0.699) 0.003^* 0.933(0.897, 0.970) < 0.001^* 0.955(0.906, 1.007) 0.088

SCS – 10.73 0.698(0.621, 0.774) < 0.001^* 1.151(1.093, 1.212) < 0.001^* 1.108(1.034, 1.18) 0.004^*

SLS – 9.05 0.508(0.425, 0.589) 0.857 1.006(0.914, 1.107) 0.904

	ROC analysis	Univariate logistic regression	Multivariate logistic regression
SRS	15.09	0.622(0.545, 0.699)	0.003^*	0.933(0.897, 0.970)	< 0.001^*	0.955(0.906, 1.007)	0.088
SCS	– 10.73	0.698(0.621, 0.774)	< 0.001^*	1.151(1.093, 1.212)	< 0.001^*	1.108(1.034, 1.18)	0.004^*
SLS	– 9.05	0.508(0.425, 0.589)	0.857	1.006(0.914, 1.107)	0.904

SRS, segmental radial strain; SCS, segmental circumferential strain; SLS, segmental longitudinal strain. *p < 0.001.

Fig. 4

Receiver-operator characteristic curves with corresponding area under the curves of SRS and SCS for detecting infarcted segments. The AUCs for SRS and SCS detecting infarcted segments were 0.622 and 0.698, respectively. SRS, segmental radial strain; SCS, segmental circumferential strain.

3.5 Intra- and inter-observer reproducibility

The global strain showed higher ICC of variability than the segmental strain. The ICC of radial strain values was slightly low (0.70). SLS had slightly higher ICC compared with SRS and SCS (0.76 –0.87 vs 0.70 –0.82, 0.75 –0.81) (details are shown in the Table 6).

Table 6
Intra- and inter-observer variability of strain values

Intra-observer Inter-observer

ICC 95% CI ICC 95% CI

GRS 0.91 0.79 – 0.96 0.70 0.50 – 0.89

GCS 0.94 0.85 – 0.97 0.76 0.47 – 0.91

GLS 0.93 0.83 – 0.97 0.93 0.80 – 0.97

SRS 0.82 0.78 – 0.85 0.70 0.63 – 0.75

SCS 0.81 0.76 – 0.84 0.75 0.69 – 0.80

SLS 0.87 0.83 – 0.89 0.76 0.71 – 0.80

	Intra-observer	Inter-observer
GRS	0.91	0.79 – 0.96	0.70	0.50 – 0.89
GCS	0.94	0.85 – 0.97	0.76	0.47 – 0.91
GLS	0.93	0.83 – 0.97	0.93	0.80 – 0.97
SRS	0.82	0.78 – 0.85	0.70	0.63 – 0.75
SCS	0.81	0.76 – 0.84	0.75	0.69 – 0.80
SLS	0.87	0.83 – 0.89	0.76	0.71 – 0.80

GRS, global radial strain; GCS, global circumferential strain; GLS, global longitudinal strain; SRS, segmental radial strain; SCS, segmental circumferential strain; SLS, segmental longitudinal strain; ICC, intra-class correlation coefficient.

4 Discussion

The major findings of this study were as follows: (1) CCTA-derived segmental strains were reduced in ischemic and infarcted segments compared with normal myocardial segments respectively; (2) Segmental radial and circumferential strains were furtherly reduced in infarcted segments compared with ischemic segments, and (3) the segments with circumferential strain below –10.73% could be infarcted segments.

At patient level comparison, neither cardiac function except LVMi nor global strain parameters showed differences among normal, ischemic and infarcted patients. Previous studies have shown that the global strain values were significantly lower in patients with obstructive coronary heart disease [19]. In the present study, the average global strain value was lower in the ischemic and infarcted group, but not at a statistically significant level. This may be related to the small sample size, as well as to the overall mild degree of ischemia in the ischemic patients or less segments involved in the infarcted patients. Additionally, abnormal LV systolic function occurs further along in CAD pathway, while patients enrolled in this study had preserved LVEF [20].

At segment level comparison, CT-derived SRS and SCS values were lower in segments with reduced perfusion. SCS and SRS derived from CMR were able to differentiate ischemic from non ischemic segments according to previous studies [7]. The conclusion was corroborated in this study. SCS was independently associated with infarcted segments. SCS could be superior to SRS in discriminating infarcted from non-infarcted myocardium. Everaars et al. found that SCS was the most accurate tool for quantifying regional function compared with SRS and wall thickening [21]. Noriko et al. also demonstrated that the SCS assessed by CMR had superior discrimination ability of detecting infarcted segments with the AUC = 0.82 [22]. The proposed SCS cut-off value for diagnosing infarcted segments in this study is –10.73% with AUC = 0.698. A possible reason why AUC for detecting infarcted segments was lower in our study could be that the temporal resolution and soft tissue resolution of CCTA were lower than CMR. On the other hand, CMR is still not as widely used as CCTA due to its technical complexity and long examination time. Stress CT MPI+CCTA can provide coronary anatomy and function assessment, while CCTA-derived strain can provide myocardial motion assessment. We support that a quantitative strain analysis from the rest CCTA is a useful tool in practice to help detecting CAD, especially for the patients who are contraindicated to receive stress inducing agents. Moreover, CCTA-derived SCS may be useful for predicting the necessity of further tests such as CT-MPI, SPECT or coronary angiography [23].

Longitudinal strain is a stable and reproducible parameter in many previous literature studies [24, 25], which is consistent with the findings of our study. However, it opposes studies stating that circumferential strain had higher reproducibility [26, 27]. Firstly, different software programs such as Circle CVI42 and TomTec Arena may influence the reproducibility. Secondly, strain parameters were calculated by feature tracking based on the 3D results in this study. The 3D feature tracking considers the short- and long-axis data simultaneously for strain calculation. Previouse study also showed that 3D feature tracking presented worse intra-observer reproducibility in radial and circumferential, comparable reproducibility in longitudinal tracking compared with 2D feature tracking [28].

This study has several limitations. First, this study was a retrospective design and a small data sample was collected from a single center, limiting the generalization ability of the conclusions. Second, our study showed that normal GLS was –6.58% based on CT images, which was lower than the value based on CMR [29]. Further studies with larger study cohorts compared with CMR are required to confirm the detection value of CT-derived strain analysis. Additionally, the best cut-off value for predicting ischemia or infarcted myocardium remains a controversial subject. Last, radiation exposure for retrospective ECG-gated cardiac CT was relatively high. However, tube-current was adjusted during R-R interval to reduce the radiation dose in this study.

In conclusion, it is feasible for CCTA-derived strain to detect regional myocardial dysfunction in coronary artery disease (CAD) patients with preserved LVEF. Segmental radial and circumferential strain have the potential ability to distinguish myocardial ischemia from infarction, and normal from ischemic myocardium.

Funding information

This study received funding from Research Foundation of Jiangsu Medical Association under grant of SYH-3201150-0006 (Yi Xu).

References

Lozano

, Naghavi

, Foreman

, et al., Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010, Lancet 380 (2012), 2095–2128.

Yang

, Zhou

, Zhang

, et al., Meta-analysis: diagnostic accuracy of coronary CT angiography with prospective ECG gating based on step-and-shoot, flash and volume modes for detection of coronary artery disease, European Radiology 24 (2014), 2345–2352.

Van Rosendael

A.R.

, Shaw

L.J.

, Xie

J.X.

, et al., Superior risk stratification with coronary computed tomography angiography using a comprehensive Atherosclerotic risk score, JACC Cardiovasc Imaging 12 (2019), 1987–1997.

Holman

E.R.

, Vliegen

H.W.

, Van Geest

R.J.D.

, et al., Quantitative analysis of regional left ventricular function after myocardial infarction in the pig assessed with cine magnetic resonance imaging, Magnetic Resonance in Medicine 34 (1995), 161–169.

Treibel

T.A.

, Rossi

, Pugliese

and Davies

L.C.

, Functional assessment of coronary artery disease by cardiac computed tomography, Expert Review of Cardiovascular Therapy 15 (2017), 657–65.

Malkasian

, Hubbard

, Dertli

, et al., Quantification of vessel specific coronary perfusion territories using minimum-cost path assignment and computed tomography angiography: Validation in a swine model, Journal of Cardiovascular Computed Tomography 12 (2018), 425–435.

Shaaban

, Tantawy

S.W.

, Elkafrawy

, et al., Multiparametric rest and dobutamine stress magnetic resonance in assessment of myocardial viability, J Magn Reson Imaging 54 (2021), 1773–1781.

Tantawy

S.W.

, Mohammad

S.A.

, Osman

A.M.

, et al., Strain analysis using feature tracking cardiac magnetic resonance (FT-CMR) in the assessment of myocardial viability in chronic ischemic patients, Int J Cardiovasc Imaging 37 (2021), 587–596.

Min

J.K.

, Shaw

L.J.

and Berman

D.S.

, The present state of coronary computed tomography angiography a process in evolution, Journal of the American College of Cardiology 55 (2010), 957–965.

10.

Tee

M.W.

, Won

, Raman

F.S.

, et al., Regional strain analysis with multidetector CT in a Swine Cardiomyopathy model: Relationship to cardiac MR tagging and myocardial fibrosis, Radiology 277 (2015), 88–94.

11.

Tavakoli

, Sahba

, Cardiac motion and strain detection using 4D CT images: comparison with tagged MRI, and echocardiography, Int J Cardiovasc Imaging 30 (2014), 175–84.

12.

, Yuan

, Yu

, et al., Prevalence of decreased myocardial blood flow in symptomatic patients with patent coronary stents: Insights from low-dose dynamic CT myocardial perfusion imaging, Korean Journal of Radiology 20 (2019), 621–630.

13.

Magalhaes

T.A.

, Kishi

, George

R.T.

, et al., Combined coronary angiography and myocardial perfusion by computed tomography in the identification of flow-limiting stenosis - The CORE320 study: An integrated analysis of CT coronary angiography and myocardial perfusion, J Cardiovasc Comput Tomogr 9 (2015), 438–445.

14.

Bamberg

, Klotz

, Flohr

, et al., Dynamic myocardial stress perfusion imaging using fast dual-source CT with alternating table positions: initial experience, European Radiology 20 (2010), 1168–1173.

15.

Cerqueira

M.D.

, Weissman

N.J.

, Dilsizian

, et al., Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association, Int J Cardiovasc Imaging 18 (2002), 539–42.

16.

Tanabe

, Kido

, Uetani

, et al., Differentiation of myocardial ischemia and infarction assessed by dynamic computed tomography perfusion imaging and comparison with cardiac magnetic resonance and single-photon emission computed tomography, European Radiology 26 (2016), 3790–3801.

17.

Chen

, Zhang

, Jiang

, et al., Coronary microvascular dysfunction: An important interpretation on the clinical significance of transient ischemic dilation of the left ventricle on myocardial perfusion imaging, J Xray Sci Technol 29 (2021), 347–360.

18.

Bamberg

, Marcus

R.P.

, Becker

, et al., Dynamic myocardial CT perfusion imaging for evaluation of myocardial ischemia as determined by MR imaging, JACC Cardiovasc Imaging 7(3) (2014), 267–277.

19.

Ivan

, Biljana

, Radosava

, et al., Visual assessment vs. strain imaging for the detection of critical stenosis of the left anterior descending coronary artery in patients without a history of myocardial infarction, Eur Heart J Cardiovasc Imaging 16 (2015), 402–409.

20.

Konerman

M.C.

, Greenberg

J.C.

, Kolias

T.J.

, et al., Reduced myocardial flow reserve is associated with diastolic dysfunction and decreased left atrial strain in patients with normal ejection fraction and epicardial perfusion, J Card Fail 24 (2018), 90–100.

21.

Henk

, Marco

, Pierre

, et al., Strain analysis is superior to wall thickening in discriminating between infarcted myocardium with and without microvascular obstruction, Eur Radiol 28 (2018), 5171–5181.

22.

Oyama-Manabe

, Ishimori

, Sugimori

, et al., Identification and further differentiation of subendocardial and transmural myocardial infarction by fast strain-encoded (SENC) magnetic resonance imaging at 3.0 Tesla, Eur Radiol 21 (2011), 2362–2368.

23.

Peled

, Lamash

, Carasso

, et al., Automated 4-dimensional regional myocardial strain evaluation using cardiac computed tomography, Int J Cardiovasc Imaging 36 (2020), 149–159.

24.

Tanabe

, Kido

, Kurata

, et al., Three-dimensional maximum principal strain using cardiac computed tomography for identification of myocardial infarction, European Radiology 27 (2017), 1667–1675.

25.

, Zhang

P.Y.

, Ran

, et al., Evaluation of left ventricular myocardial mechanics by three-dimensional speckle tracking echocardiography in the patients with different graded coronary artery stenosis, Int J Cardiovasc Imaging 33 (2017), 1513–1520.

26.

Augustine

, Lewandowski

A.J.

, Lazdam

, et al., Global and regional left ventricular myocardial deformation measures by magnetic resonance feature tracking in healthy volunteers: comparison with tagging and relevance of gender, J Cardiovasc Magn Reson 15 (2013), 8.

27.

Morton

, Schuster

, Jogiya

, et al., Inter-study reproducibility of cardiovascular magnetic resonance myocardial feature tracking, J Cardiovasc Magn Reson 14 (2012), 43.

28.

, Paul

, Li

, et al., Left ventricular myocardial strain quantification with two- and three-dimensional cardiovascular magnetic resonance based tissue tracking, Quant Imaging Med Surg 11 (2021), 1421–1436.

29.

Backhaus

S.J.

, Metschies

, Zieschang

, et al., Head-to-head comparison of cardiovascular MR feature tracking cine versus acquisition-based deformation strain imaging using myocardial tagging and strain encoding, Magn Reson Med 85 (2021), 357–368.

	Intra-observer		Inter-observer
	ICC	95% CI	ICC	95% CI
GRS	0.91	0.79 – 0.96	0.70	0.50 – 0.89
GCS	0.94	0.85 – 0.97	0.76	0.47 – 0.91
GLS	0.93	0.83 – 0.97	0.93	0.80 – 0.97
SRS	0.82	0.78 – 0.85	0.70	0.63 – 0.75
SCS	0.81	0.76 – 0.84	0.75	0.69 – 0.80
SLS	0.87	0.83 – 0.89	0.76	0.71 – 0.80